2011 Poster Sessions : Blind Portfolio Auctions

Student Name : Michael Padilla
Advisor : Benjamin Van Roy
Research Areas: Information Systems
As much as 12% of the daily volume on the New York Stock Exchange involves sales by institutional investors to brokers through blind portfolio auctions. Such transactions typically take the form of a first-price sealed-bid auction in which the seller engages a few potential brokers and provides limited information about the portfolio being sold. Uncertainty about the portfolio contents reduces bids, effectively increasing the transaction cost paid by the seller. We consider the use of a trusted intermediary, or equivalent cryptographic protocol, to reduce transaction costs. In particular, we propose a mechanism through which each party provides relevant private information to an intermediary who ultimately reveals only the portfolio contents and price paid, and only to the seller and winning bidder. Through analysis of a game-theoretic model, we demonstrate substantial potential benefits to sellers. For example, under reasonable assumptions a seller can reduce expected transaction costs by more than 10%. We will also present related ongoing work, including empirical studies using data provided by market practitioners.

Michael Padilla is a PhD candidate in the Information Systems Laboratory in the Department of Electrical Engineering at Stanford University, working with Professor Benjamin Van Roy. He has prior degrees in electrical engineering from MIT (S.M.) and UCSD (B.S.) and worked for Morgan Stanley in Tokyo prior to coming to Stanford for his PhD. His research interests are operational efficiency issues in business, computational finance and economics, multi-agent systems, and mechanism design.